1. Erasure Coding + Disaggregated Memory

    Disaggregated Memory is currently a hot topic in systems research, and distributed large-capacity memory clearly requires system-level reliability strategies. While replication has always been a default choice, with many related works, including recent ones like SWARM@SOSP’24, erasure coding is also an option. This article lists existing EC+DM works.

    2024/12/05 EC

  2. Erasure Coding NIC Offload

    About offloading erasure coding to NICs.

    2024/12/04 EC

  3. Data Movement with DMA/DSA Offloading

    About offloading memory data movement to DMA or DSA engines.

    2023/10/10 System

  4. SW Prefetch in System&DB

    Prefetch to hide memory access latency (CPU stall) What to prefetch When to prefetch Where to place the prefetched data

    2023/10/09 System

  5. Fault Tolerance of Persistent Memory

    In this article, we will list several papers on local NVM/PM fault tolerance.

    2022/09/29 PM

  6. QoS on Persistent Memory Systems

    QoS (LB) on persistent memory systems to avoid interference.

    2022/03/12 PM

  7. RDMA+NVM remote persistence

    Problem Due to RDMA NIC implementation, RNIC doesn’t have remote persistent flush primitives. So one-sided write data from clients will write to the volatile cache on RNIC first and then RNIC directly sends ACK back before writing data to PM. As a result, a power loss will break remote data persistence easily.

    2021/10/19 PM

  8. (SC '21) LogECMem: Coupling Erasure-Coded In-memory Key-Value Stores with Parity Logging

    LogECMem uses a hybrid method of in-place update and Parity logging (PL) for parity updates.

    2021/10/11 EC

  9. (VLDB '22) PM + Learned Index

    learned index + PM. APEX: A High-Performance Learned Index on Persistent Memory[1]

    2021/09/17 PM

  10. (ATC '21, SDC' 20) Use PM as cache in industry

    Some industry works about how to utilize DRAM+PM archi as cache (from facebook and twitter).

    2021/08/01 PM